ChE Board common mistakes
ChE Board · Understand the traps and how to avoid them.
Setting up a material balance and then solving it with mole fractions where mass fractions were given (or vice versa), producing a composition that does not close on a total-mass or total-mole basis.
Candidates read the problem statement quickly, see percentages, and default to whichever basis they used last. The mix-up is silent because the arithmetic still looks reasonable; the error only surfaces when the total is checked.
What to do instead
Write the basis explicitly at the top of the solution ('basis: 100 kg feed' or 'basis: 1 mol feed') and label every stream variable with its units (kg/h, kmol/h). Convert all fractions to the chosen basis before writing any balance equation, and finish by substituting back to confirm total in = total out. If the problem gives both mass and mole data, do the conversion in a dedicated step rather than inline.
Using the arithmetic-mean temperature difference in a heat exchanger instead of the log-mean, or applying the log-mean without the correction factor F for a multipass or cross-flow unit.
The arithmetic mean is faster and is correct only when the two terminal differences are nearly equal; candidates under time pressure skip the check. Counter-current versus multipass geometry is also frequently overlooked.
What to do instead
Always compute both terminal differences first; if they differ by more than roughly 10 percent, use the log-mean. For anything other than pure counter-current or pure co-current, look up or compute F from the R and P parameters and multiply. Sanity-check that F lies between about 0.75 and 1.0; a value outside that range signals a wrong configuration or an arithmetic slip.
Treating a reactor as isothermal or as reaching equilibrium when the problem specifies adiabatic operation or gives kinetics with a residence time, and reporting a conversion that ignores the energy balance or the rate expression.
Candidates pattern-match on the reactor type (CSTR, PFR, batch) and apply the design equation alone, forgetting that an adiabatic reactor's conversion is coupled to the temperature rise through the heat of reaction.
What to do instead
Before solving, list the constraints the problem actually states: isothermal or adiabatic, equilibrium-limited or kinetically limited, constant volume or variable volume. For adiabatic cases, write the energy balance alongside the mole balance and solve them together, or use the adiabatic temperature rise to find T as a function of conversion. For kinetic cases, confirm the units of the rate constant match the order before integrating.
Applying the mechanical energy balance but dropping the friction term, or using the Bernoulli equation across a pump, valve, or fitting where head loss is significant.
Bernoulli is the most familiar form and candidates reach for it automatically; the friction head is easy to omit because it requires a friction factor, a Reynolds number, and a length.
What to do instead
Start from the full mechanical energy balance and cross out only terms that are genuinely zero (for example, no shaft work in a pipe run, or equal velocities in a constant-diameter line). Compute the Reynolds number and friction factor explicitly whenever a pipe length, fitting, or valve appears. For pump problems, keep the shaft-work term and verify the sign convention: work added to the fluid is positive on the left side.
Selecting a unit operation or equipment size from a shortcut rule of thumb without checking the governing constraint, such as choosing a distillation reflux ratio or a tray count without verifying the pinch or the flooding limit.
Shortcut methods (Fenske-Underwood-Gilliland, Souders-Brown) are taught as fast design tools and candidates apply the result as final rather than as a starting estimate. The physical limit that actually governs the design is then never tested.
What to do instead
Use shortcut methods only to bracket the answer, then check the binding constraint: minimum reflux and pinch composition for distillation, flooding velocity and downcomer capacity for trays, pressure drop and minimum fluidization for packed or fluidized beds. State in the answer which constraint governs and show that the proposed design sits below it with a margin.
Mixing gauge and absolute pressure, or using the wrong reference temperature in a thermodynamic property calculation, so enthalpies and vapor pressures are offset by a constant.
Property tables and steam data use different reference states, and problem statements often give gauge readings. The resulting error is a uniform shift that does not look obviously wrong.
What to do instead
Convert every pressure to absolute immediately and note the conversion in the working. When using enthalpy or entropy data, confirm the reference state of the table and use one consistent source throughout; do not mix tabulated steam enthalpies with values computed from a heat-capacity integral referenced to a different zero. Check that a saturated mixture's enthalpy falls between the saturated liquid and vapor values.
Reporting a dimensionless group or correlation result without confirming unit consistency, especially in heat transfer and fluid mechanics correlations where the Reynolds, Nusselt, and Prandtl numbers must be evaluated in a single consistent unit system.
Correlations are written for a specific unit convention (for example, Dittus-Boelter with properties in SI or in a stated set of English units), and candidates plug in whatever numbers the problem supplies.
What to do instead
Convert all properties to one system before substitution, and write the units next to each property as it is entered. Verify that each dimensionless group comes out unitless; if a group carries units, a conversion was missed. For the heat transfer coefficient, confirm that the thermal conductivity and diameter are in matching length units before dividing.
Ignoring the control objective and tuning a PID loop by trial values rather than by identifying the process gain, time constant, and dead time from the given response data.
Candidates memorize tuning rules by name but do not extract the model parameters the rules require, so they substitute guessed values or the wrong rule for the loop type.
What to do instead
From the step-response data, read off the process gain (steady-state change divided by input change), the time constant (time to reach about 63 percent of the final change), and the dead time. Then apply the tuning rule that matches the controller mode (P, PI, or PID) and the desired response. State the assumed model form and the rule used so the reasoning is visible.
Skipping the safety and operability check on a proposed design, such as omitting relief sizing, flammability limits, or material compatibility when the problem describes a hazardous service.
Candidates treat the question as a pure calculation and stop when the number is found, forgetting that board-level problems often embed a safety or integrity requirement in the scenario.
What to do instead
After the calculation, run a short checklist: does the stream fall within flammability limits, does the vessel need relief protection, is the material compatible with the service, and are there runaway or exotherm risks. If any apply, size or specify the safeguard and include it in the answer. Even a one-line statement of the safeguard earns credit and demonstrates engineering judgment.
Presenting a final numeric answer with no units, no significant figures discipline, and no reasonableness check against physical bounds.
Under exam time pressure candidates stop at the number; the last-minute check is the first thing dropped, and a result that is off by orders of magnitude goes undetected.
What to do instead
Carry units through every step and box the final answer with units. Before moving on, test the result against bounds: a conversion must lie between 0 and 1, a mole fraction between 0 and 1, an efficiency below 100 percent, and a temperature above absolute zero. If the number violates a bound, recheck the basis and unit conversions before proceeding.